Tau Acetylation in Entorhinal Cortex Induces its Chronic Hippocampal Propagation and Cognitive Deficits in Mice

Abstract

Background:

Increased tau acetylation at K174, K274, K280, and K281 has been observed in the brains of Alzheimer’s disease (AD) patients or in transgenic mice, but the role of acetylation in tau propagation is elusive.

Objective:

To study the effect of tau acetylation in entorhinal cortex on tau transmission and learning and memory.

Methods:

Stereotactic brain injection, behavioral test, electrophysiological recording, immunohistochemistry, and immunofluorescence were used.

Results:

We constructed the hyperacetylation mimics of tau (AAV-Tau-4Q), the non-acetylation tau mutant (AAV-Tau-4R), and the wild-type tau (AAV-Tau-WT). By overexpressing these different tau proteins in the entorhinal cortex (EC) of 2-month-old mice, we found that overexpressing Tau-4Q in EC for 3 or 6 months (to 5 or 8 months of age) neither induces tau propagation to dentate gyrus (DG) nor glial activation in DG, nor spatial memory deficit. However, overexpressing Tau-WT and Tau-4Q in EC for 13.5 months (15.5 months of age) at 2 months promoted tau propagation respectively to granulosa and hilus of DG with glial activation, synaptic dysfunction, and memory deficit, while overexpressing Tau-4R abolished tau propagation with improved cellular pathologies and cognitive functions. Furthermore, overexpressing Tau-4Q in unilateral DG of 2-month-old mice for 8 weeks also promoted its contralateral transmission with glial activation, and mice with tau (Tau-WT, Tau-4Q, and Tau-4R) overexpression in DG showed cognitive deficits compared with the empty vector controls.

Conclusion:

Tau acetylation induces a time-dependent propagation from EC to DG, and only hippocampus but not EC tau accumulation induces cognitive deficits.

Keywords

Acetylation Alzheimer’s disease neuroglia propagation tau

INTRODUCTION

Alzheimer’s disease (AD) is the most common neurodegenerative disease characterized by extracellular deposition of amyloid-β (Aβ) and intracellular accumulation of the hyperphosphorylated tau protein. Post-translational modification of tau is a key factor affecting tau aggregation and neurodegeneration. Phosphorylation of tau is the most extensively studied [1], but phosphorylation alone does not necessarily lead to tau aggregation [2, 3]. Therefore, more and more studies have recently turned to other post-translational modifications, such as acetylation. It has been reported that the acetylation level of tau at K174 [4], K274 [5], K280 [6], and K281 [5, 7] is increased in the brains of AD patients or the transgenic mice, and the acetylated K174 is co-stained with the misfolded tau [4]. Tau acetylation affects tau degradation [4–6 , 9], resulting in excessive tau aggregation. Thus, the acetylation of tau is closely related to the abnormal aggregation of tau.

During the progression of AD, tau started to form aggregates or neurofibrillary tangles in the entorhinal cortex, and then spread through limbic hippocampi to the hippocampal subsets including dentate gyrus (DG), and finally to the whole brain [10, 11]. This has been served as a gold standard for the classification of AD pathology [12]. Currently, it is not clearly understood what may be the molecular mechanisms underlying this pathological tau propagation during the chronic course of AD. It is reported that microglial activation can promote tau spreading [13]. Microglia may take up tau aggregates through endocytosis and exhalation [14 –16], which generally causes further accumulation of tau [17]. Tau may also be transmitted via synapses in the form of exosomes, resulting in radiating propagation of tau pathology from the heavy pathological areas to the mild pathological subsets [17, 18]. In addition, studies also suggest that tau aggregation is a fundamental of tau propagation among various promoting factors [3 , 19–21].

In the present study, we studied the role of tau acetylation in its propagation. We found that acetylation of tau at AD-related sites (K174, K274, K280, and K281) promotes tau propagation from the entorhinal cortex to the hippocampal dentate gyrus in a time-dependent manner with glial activation. Interestingly, tau accumulation in EC alone does not induce cognitive impairment, while hippocampal propagation of tau causes cognitive deficits.

MATERIALS AND METHODS

Animals

Wild-type C57BL/6 mice (2 months old, male) were purchased from Beijing Huafukang Company. 3×Tg AD mice (stock number 004807) and APP/PS1 mice (stock number 34829-JAX) were purchased from Jackson Laboratory (USA) in accordance with the animal and plant inspection and quarantine law of the People’s Republic of China. The experimental animals were managed in strict accordance with the regulations on the administration of experimental animals in China. The mice were kept in the same environment, with temperature control of 23°C, sufficient water, free intake, and a circadian rhythm of 12:12×h. All animal experiments were conducted in accordance with Policies on the Use of Animals and Humans in Neuroscience Research.

Stereotactic brain injection

The mice were anesthetized by isoflurane and infused stereotaxically (KOPF brain stereotactic locator, Germany) into EC or DG subset with non-fused eGFP virus constructs, including AAV-syn-eGFP-2a-tau-4Q (Lys mutated to Gln at K174, K274, K280 and K281)-3flag (termed as AAV-Tau-4Q) and the controls, AAV-syn-eGFP-2a-MCS-3flag (empty vector, termed as AAV-Vec), AAV-syn-eGFP-2a-tau-WT (wild-type tau)-3flag (termed as AAV-Tau-WT), and the non-acetylation AAV-syn-eGFP-2a-tau-4R (Lys mutated to Arg at K174, K274, K280, and K281)-3flag (termed as AAV-Tau-4R). The coordinates for the injection were as follows: DG (anterior-posterior: –1.9 mm; mediolateral: –0.7 mm; dorsoventral: –2.2 mm from bregma and dura, flat skull), EC (anterior-posterior: –4.8 mm; mediolateral: ±2.8 mm; dorsoventral: –3.5 mm from bregma and dura, flat skull). We used different cohorts of mice for three surviving periods, i.e., 3 months, 6 months, and 13.5 months after brain infusion of the virus constructs. For the age-matched controls, the mice were sham operated and same volume of artificial cerebrospinal fluid was infused. The behavioral tests were carried out at 3 months, 6 months, or 13.5 months after the injection, and then the mice were sacrificed for the biochemical measurements.

Open field (OF) test

The mice were placed in an open field of 50×50×50 cm, and the bottom of the box was divided into 9 parts, the central area occupied 50%, and the edges and four corners were the other 50%. The central stagnation time of the mice within 5 min was calculated by a video tracking system (Chengdu Taimeng Software Co. Lid, China). Before each behavioral experiment was performed on each mouse, the test bench was sprayed with 75% alcohol and wiped clean.

Elevated plus maze (EPM) test

The equipment was two cross platforms, 60 cm long and 10 cm wide, and two arms were open arms and two arms were closed arms. The mice were placed in the cross position in the middle of the elevated cross, and the open arm residence time of the mice was recorded within 5 min by a video tracking system (Chengdu Taimeng Software Co. Lid, China). Before each behavioral experiment was performed on each mouse, the test bench was sprayed with 75% alcohol and wiped clean.

Novel object recognition (NOR) test

The NOR test was carried by following a previous report [22]. The mice were placed in an open field of 50×50×50 cm, with two objects placed in the corners of open field, to make mice explore for 5 min. After 24 h, one of the objects was changed and the mice were allowed to move freely. When the mice came within 3 cm of the object, the preference toward the novel object within 5 min were recorded to test the memory ability of mice by a video tracking system (Chengdu Taimeng Software Co. Lid, China). There is a camera to record the location of the distance and time in the center of the top. Before each behavioral experiment was performed on each mouse, the test bench was sprayed with 75% alcohol and wiped clean.

Morris water maze (MWM) test

The MWM was a circular pool with a diameter of 120 cm and a wall height of 60 cm. The spatial learning and memory capacities of the mice were recorded by a video tracking system recorded by a camera set above the pool center (Chengdu Taimeng Software Co. Lid, China). The mice were allowed to acclimate for a week before the experiment. During the learning trial, the mice were gently put into the pool facing the wall and allowed to swim freely. The time for the mice to find the platform was recorded as the latency and the mice were allowed to stay on the platform for at most 30 s. If the mouse failed to find the platform in 60 s, the experimenter gently guided the mouse to the platform and allowed it to stay there for at most 30 s. The learning trial was repeated 3 times a day for 6 or 7 days, and the interval between the two training sessions of the same mouse was at least 30 min.

Spatial memory was tested 2 days after the last training trial by removing the platform. The time (s) used to reach the previous platform site was recorded as latency, and the times crossed the platform area within 60 s was recorded as target platform crossings.

Fear conditioning (FC) test

The equipment is a 20×20×40 cm box with a soundproof door. The bottom of the box can give an electric shock. The camera above the center of the box can record the movement track, distance, and time. The mice were put into the cage to adapt for 3 min, then 2 s foot shock (100 mA with an interval of 1 min) was repeatedly given for three times, then the mice were put back into the cage. After 24 h, the mice were put into the box and the freezing time of the mice was recorded within 3 min by a video tracking system (Chengdu Taimeng Software Co. Lid, China). Freezing is defined when the mouse’s body deviated from the original position within 10%, and the statistical results were obtained. Before each behavioral experiment was performed on each mouse, the test bench was sprayed with 75% alcohol and wiped.

Electrophysiological recording on brain slices

The mice were anesthetized with isoflurane, and the forelimbs were gently clipped with forceps. After confirming no retraction, the mice were sacrificed and the brains were taken out. The brain slices were cut (300 μm) in ice-cold artificial cerebrospinal fluid (ACSF) containing (in mM): NaCl 124; KCl 3.0; MgCl₂ 1.0; CaCl₂ 2.0; NaH₂PO₄ 1.25; NaHCO₃ 26; glucose 10; saturated with 95% O₂, 5% CO₂ (pH 7.4). Then, slices were incubated at room temperature for 30 min in ACSF. Voltage signals were acquired using the MED64 System (Alpha MED Sciences). The fEPSPs were obtained by electrically stimulating the DG region. Stimulation intensity was adjusted to evoke fEPSP amplitudes that were 40% of maximal size. Long-term potentiation (LTP) was induced by applying one or three train(s) of high-frequency stimulation (100 Hz, 1 s duration at test strength).

Immunohistochemistry and immunofluorescence

The isoflurane anesthetized mice were fixed with abdomen up, and then perfused sequentially through heart injection with normal saline (NS) and 4% paraformaldehyde (PFA). After immersion fixed in 4% PFA for another 24 h, the brains were dehydrated in 30% sucrose-PBS for 48 h, and were cut into 30μm coronal or horizontal sections. The immunohistochemistry (IHC) and Immunofluorescence (IF) experiments were carried out according to the established method [23]. The slices were incubated with primary antibody (IBA1, Wako, 1:500; HT7, ThermoFisher Scientific, 1:400; GFAP, Abcam, 1:500; Hoechst, Sigma, 1:1000; Tau (Acetyl Lys174), Immunoway, 1:200; Tau (Acetyl Lys280), AnaSpec, 1:50). The images were taken using an ordinary optical microscope for IHC (Nikon, Tokyo, Japan) or using a confocal microscope for IF (Carl Zeiss LSM800).

Statistical analyses

The data were presented as mean±SEM. One-way or two-way ANOVA with Turkey post hoc tests and GraphPad Prism 8 software were used for multi-group comparison. The statistically significance was set at p < 0.05.

RESULTS

Overexpressing Tau-4Q in EC for 3 or 6 months neither induces tau propagation to DG nor glial activation in DG, nor spatial memory deficit

To explore the role of tau acetylation in its transmission, we first measured the acetylation level of tau in the EC subset, the recognized starting site for abnormal tau propagation observed in AD patients [12 , 25]. By immunohistochemical staining using tau-K174 and tau-K280 antibodies, we observed that the level of acetylated tau was significantly increased in EC subset of 6-month-old APP/PS1 and 3×Tg AD mice (Fig. 1A, B). Among many potential acetylation sites in tau proteins, K174, K274, K280, and K281 have been detected in the human AD brains (K174, K274, K281) or AD transgenic mice (K280). Therefore, we construct acetylation mimics AAV-Tau-4Q (Lys mutated to Gln at K174, K274, K280, and K281) and the controls, including AAV-Vec (empty vector), AAV-Tau-WT (wild-type tau),and the non-acetylation AAV-Tau-4R (Lys mutated to Arg at K174, K274, K280, and K281). We infused different virus constructs respectively into the EC subset of 2-month-old C57 mice. After three months (5 months of age), the expression of tau in EC subset was confirmed by immunofluorescence staining using HT7, an antibody specifically reacts with human tau proteins (Fig. 1C, Supplementary Figure 1A). Three months after tau overexpression, neither tau propagation to DG subset nor glial cell activation in the DG subset was detected (Fig. 1D–F). In EC subset, microglial activation was detected with no significant change of astrocytes at this age (Supplementary Figure 1B, C). Meanwhile, no significant cognitive impairments were detected by open field (OF), elevated plus maze (EPM), novel object recognition (NOR), Morris water maze (MWM), and fear conditioning (FC) (Fig. 1H–K, Supplementary Figure 2). Then, we reared the mice for 6 months after stereotaxic infusion of different AAV-tau constructs into the EC subset (8 months of age). Consistent expression of the virus in the EC subset was confirmed (Supplementary Figures 3A and 4A). However, we still did not detect significant tau propagation to DG (Fig. 1G, Supplementary Figure 4B) or glial cells activation in DG (Fig. 1G, Supplementary Figure 4C, D), though microglial and astrocyte activation in EC subset was shown (Supplementary Figure 3B, C). By behavior tests, no significant learning and memory deficits were shown in the mice (Fig. 1L–O, Supplementary Figure 5). These data together demonstrate that overexpressing tau, including wild type and the four lysine-sites mutated tau, in EC subset for 3 or 6 months neither induces tau propagation to the hippocampal DG subset nor cognitive impairment; furthermore, overexpressing hyperacetylated Tau-4Q in EC for 3 and 6 months does not promote tau propagation.

Fig.1

Overexpressing Tau-4Q in EC subset for 3 months or 6 months neither induce tau propagation to DG nor glial activation in DG nor spatial memory deficits in mice. A, B) In 6-month-old APP/PS1 and 3×Tg mice, the immunoreactivity of tau-K174 or tau-K280 in EC subset was increased compared with the age-matched control mice. Scale bar, 100μm in A and 500μm in B. N = 3 in each group, two-tail t-test. Data were expressed as mean±SEM, *p< 0.05, **p < 0.01. C–O) The eGFP-non-fused virus constructs including AAV-Tau-WT, AAV-Tau-4Q, AAV-Tau-4R, or the empty vector (AAV-Vec) was infused into EC subset of 2-month-old C57 mice. After 3 or 6 months, the following tests were carried out. C) Tau expression in EC subset was confirmed by immunofluorescence staining using HT7 that specifically probes human tau proteins after 3 months. Scale bar, 500μm. D) No HT7 signal in DG subset after 3 months. Scale bar, 10μm. E, F) No significant difference of IBA1 or GFAP signals in DG subset after 3 months. Scale bar, 50μm. G) No HT7 signal or no significant difference of IBA1 or GFAP signals in DG after 6 months. Scale bar, 500μm, 10μm, 50μm from left to right. H) No difference during 7 days learning trial by MWM test (3 months). Data were presented as mean±SEM (N = 10∼14 each group, repeated measures two-way ANOVA). I–K) No difference during probe trial measured at day 9 after removed the platform (3 months): the escape latency to find the hidden platform (I), the target platform crossings (J), and the times in the target quadrant (K). Data were presented as mean±SEM (N = 10∼14 each group, one-way ANOVA). L) No difference during 7 days learning trial by MWM test (6 months). Data were presented as mean±SEM (N = 10∼13 each group, repeated measures two-way ANOVA). M–O) No difference during probe trial measured at day 9 after removed the platform (6 months): the escape latency to find the hidden platform (M), the target platform crossings (N), and the times in the target quadrant (O). Data were presented as mean±SEM (N = 10∼13 each group, one-way ANOVA).

Overexpressing Tau-4Q in EC for 13.5 months promotes tau propagation to DG subset with glial activation and memory deficit

We further studied whether tau acetylation promotes its transmission by reared the mice to 13.5 months after EC infusion of different tau constructs at 2 months of age. By immunofluorescence staining, we observed that expressing Tau-WT induced tau propagation to the granulosa of DG, while expressing Tau-4Q promoted tau propagation to the inner hilus of DG (Fig. 2A). Similar results were also shown by immunohistochemical staining (Fig. 2B). On the other hand, low positive HT7 signal was detected in DG subset after overexpressing Tau-4R (Fig. 2) compared with Tau-4Q. These data demonstrated that the hyperacetylated tau propagated much faster than the wild-type tau, while acetylation sites mutation abolished the propagation. Together with the observations shown in Fig. 1, it is suggested that tau acetylation can promote its propagation in a time-dependent manner, which exactly fits the chronic feature of AD pathologies.

Fig.2

Overexpressing Tau-4Q in EC for 13.5 months promotes tau propagation to DG subset. The eGFP-non-fused virus constructs including AAV-Tau-WT, AAV-Tau-4Q, AAV-Tau-4R, or the empty vector (AAV-Vec) was infused into EC subset of 2-month-old C57 mice for 13.5 months. A) Propagation of Tau-WT and Tau-4Q respectively to granulosa and the inner hilus of DG subset measured by HT7 immunofluorescence staining. Scale bar, 100μm. B) The quantitative analyses of panel A. Data were presented as mean±SEM, N = 3 (hippocampal slices from 3 mice in each group), one-way ANOVA. **p < 0.01, versus Vec; ^&&p < 0.01, versus Tau-WT; ^##p < 0.01 versus Tau-4R. C) Propagation of Tau-WT and Tau-4Q respectively to granulosa and the inner hilus of DG subset measured by HT7 immunohistochemical staining (13.5 months). Scale bar, 20μm. D) The quantitative analyses of panel C. Data were presented as mean±SEM, N = 3, one-way ANOVA. **p < 0.01, versus Vec; ^&&p < 0.01, versus Tau-WT; ^##p < 0.01 versus Tau-4R.

By immunofluorescence co-staining of HT7 with GFP, we confirmed the expression of human tau in the EC subset for 13.5 months (15.5 months of age) after infusion of the virus constructs (Fig. 3A). We also observed that overexpressing Tau-WT and Tau-4Q but not Tau-4R induced microglia and astrocyte activation in both EC (Fig. 3B, C) and DG subsets (Fig. 3D, E), and the more significant effect was shown in Tau-4Q group (Fig. 3B–E). By NOR test, we observed that the mice with overexpression of Tau-WT and Tau-4Q but not Tau-4R showed reduced ability to recognize the novel object (Fig. 3F) with no changes on motor function (Fig. 3G). These data suggest that acetylation-promoted tau propagation from EC to hippocampus induces glial activation and memory deficit of the mice.

Fig.3

Overexpressing Tau-4Q in mouse EC for 13.5 months induces glial activation and cognitive deficits. The eGFP-non-fused virus constructs including AAV-Tau-WT, AAV-Tau-4Q, AAV-Tau-4R, or the empty vector (AAV-Vec) was infused into EC subset of 2-month-old C57 mice for 13.5 months. A) Tau expression in EC subset was confirmed by immunofluorescence staining using HT7 that specifically probes human tau proteins. Scale bar, 50μm. B, C) Activation of microglia (IBA1) and astrocyte (GFAP) was detected in EC subset of Tau-WT and Tau-4Q group, and most significant activation was shown in Tau-4Q group. Scale bar, 20μm. D, E) Activation of microglia and astrocyte was detected in DG subset of Tau-WT and Tau-4Q group, and most significant activation was shown in Tau-4Q group. Scale bar, 50μm (IBA1) and 20μm (astrocyte). F, G) An impaired novel object recognition preference was shown in Tau-WT and Tau-4Q groups, and more significant deficit was detected in Tau-4Q group. No significant difference of the motor function was detected in different groups. Data were presented as mean±SEM (N = 8∼10 each group, one-way ANOVA). *p < 0.05, **p < 0.01, versus Vec; ^##p < 0.01 versus Tau-4R.

Overexpressing Tau-4Q in hippocampal DG promotes its contralateral transmission and glial activation

From the above results, it seems that only those tau proteins transmitted to hippocampus cause memory impairment. Therefore, we further explored the role of tau acetylation in its propagation and the toxicity by overexpressing different AAV-tau constructs in the unilateral hippocampus of 2-month-old C57 mice. After 8 weeks, the expression of human tau in the infused ipsilateral hippocampus was confirmed by co-immunofluorescence staining of HT7 with GFP (Fig. 4A, B). The HT7-positive signals were also detected in the contralateral hippocampus of the mice with overexpression of Tau-WT and Tau-4Q, and the HT7-positive cell number was much higher in Tau-4Q mice than the Tau-WT mice (Fig. 4C). On the contrary, overexpressing Tau-4R abolished tau propagation to the contralateral hippocampus (Fig. 4C). These data further confirm that acetylation can promote tau propagation.

Fig.4

Overexpressing Tau-4Q in DG subset promotes its contralateral transmission. The eGFP-non-fused virus constructs including AAV-Tau-WT, AAV-Tau-4Q, AAV-Tau-4R, or the empty vector (AAV-Vec) was infused into the unilateral hippocampal DG subset of 2-month-old C57 mice for 8 weeks. A) Images showing injection site at ipsilateral hippocampal DG, and the whole picture of ipsilateral and contralateral DG subsets of the brain coronal section. Scale bar, 1000μm. B) The enlarged images confirmed expression of all tau constructs in ipsilateral hippocampal DG subset measured by HT7 immunofluorescence staining. Scale bar, 50μm. C) The quantitative analyses of panel B. Data were presented as mean±SEM, N = 3 (hippocampal slices from 3 mice in each group), one-way ANOVA. **p < 0.01, versus Vec. D) The enlarged images showing propagation of minor Tau-WT and prominent Tau-4Q to the contralateral DG subset measured by HT7 immunofluorescence staining. Scale bar, 50μm. E) The quantitative analyses of panel D. Data were presented as mean±SEM, N = 3, one-way ANOVA. **p < 0.01, versus Vec; ^&&p < 0.01, versus Tau-WT; ^##p < 0.01 versus Tau-4R.

By immunofluorescence staining, we observed that overexpressing Tau-4Q induced very significant activation of microglial and astrocyte in both ipsilateral (Fig. 5A, B, D, E) and the contralateral sites (Fig. 5A, C, D, F) of the hippocampal DG, though fewer microglial activation was also shown in the ipsilateral hippocampus of Tau-WT group (Fig. 5B). These data suggest that tau acetylation promotes glial activation. We also measured LTP in the acute brain sections. The results showed that the EPSP slope was decreased in all three groups with overexpression of different types of tau proteins, and the decrease was most significant in Tau-4Q group than the Tau-WT group, and no difference was shown between Tau-WT and the Tau-4R groups (Fig. 5G, H). These data suggest that tau acetylation impairs synaptic function.

Fig.5

Overexpressing Tau-4Q in DG subset deteriorates wild-type tau-induced glial activation and the deficits in synaptic function. The eGFP-non-fused virus constructs including AAV-Tau-WT, AAV-Tau-4Q, AAV-Tau-4R, or the empty vector (AAV-Vec) was infused into the unilateral hippocampal DG subset of 2-month-old C57 mice for 8 weeks. A–C) Images and the quantitative analyses show minor or prominent microglial activation at ipsilateral and the contralateral hippocampal DG in Tau-WT and Tau-4Q. Data were presented as mean±SEM, N = 3 hippocampal slices from 3 mice in each group, one-way ANOVA. **p< 0.01, versus Vec; ^&p< 0.05, ^&&p < 0.01, versus Tau-WT; ^##p < 0.01 versus Tau-4R. Scale bar, 50μm (DG) and 1000μm (hippocampus). D–F) Images and quantitative analyses show prominent astrocytes activation at ipsilateral and the contralateral hippocampal DG in Tau-4Q group. Data were presented as mean±SEM, N = 3 hippocampal slices from 3 mice in each group, one-way ANOVA. **p < 0.01, versus Vec; ^&p < 0.05, ^&&p < 0.01, versus Tau-WT; ^##p < 0.01 versus Tau-4R. Scale bar, 50μm (DG) and 1000μm (hippocampus). G, H) The EPSP slope in Tau-WT and Tau-4Q groups were decreased, and the reduction was more significant in Tau-4Q group measured by brain slice electrophysiological recording. The quantification was done by collected the data at last 10 min. N = 6 hippocampal slices from 3 mice in each group, one-way ANOVA. Data were expressed as mean±SEM, **p < 0.01, versus Vec; ^&&p < 0.01 versus Tau-WT; ^##p < 0.01 versus Tau-4R.

Overexpressing Tau-4Q in hippocampal DG does not significantly deteriorate cognitive deficits

We further explored the effects of hippocampal overexpression of the acetylated tau on cognitive functions of the mice. The 2-month-old C57 mice were stereotaxically infused with different AAV-tau constructs for 8 weeks, and then the cognitive functions of the mice were tested sequentially by OF, EPM NOR, MWM, and FC paradigms (Fig. 6A). In the OF and EPM tests, all the mice with tau overexpressing (including Tau-WT, Tau-4Q, and Tau-4R) showed a significantly reduced stay in the center (Fig. 6B) or a significantly reduced time in the open arm (Fig. 6C), suggesting that ipsilateral hippocampal overexpressing tau for 8 weeks induced depression-like behaviors in the mice. In NOR test, both Tau-WT and Tau-4Q mice showed a significantly reduced preference toward novel object, and a more profound reduction was shown in Tau-4Q mice while overexpressing Tau-4R restored the ability of the mice in recognizing the novel object (Fig. 6D). These data suggest that overexpressing the acetylated tau at ipsilateral hippocampal promotes memory deficit in recognizing the new object. The mice did not show any difference in moving distance during NOR tests (Fig. 6E), which excluded any potential motor dysfunction of the mice. Spatial learning and memory were tested by using MWM. During the 6 days learning trials, the mice with overexpression of Tau-WT and Tau-4Q but not Tau-4R showed significantly increased latency to find the hidden platform (Fig. 6F), suggesting spatial learning impairment by Tau-WT and Tau-4Q. On day 8, the spatial memory was tested by removed the hidden platform. The results showed that Tau-WT and Tau-4Q mice spent longer time to reach the previous platform location with fewer crossings times in the target quadrant than the mice expressing the empty vector or Tau-4R (Fig. 6G–I). No difference in travel distance was detected in four groups of mice during MWM test (data not shown). These data indicate that expressing acetylated tau in the ipsilateral hippocampus induces spatial memory deficits. Finally, we used the FC paradigm to test the fear memory of the mice. The results showed that the freezing time was decreased in Tau-WT and Tau-4Q mice compared to the mice expressing empty vector while overexpressing Tau-4R abolished the effects of Tau-WT and Tau-4Q (Fig. 6J), which further confirmed that acetylation of tau promotes memory deficits.

Fig.6

Overexpressing Tau-4Q in hippocampal DG does not significantly aggravate Tau-WT-induced cognitive deficits. The eGFP-non-fused virus constructs including AAV-Tau-WT, AAV-Tau-4Q, AAV-Tau-4R, or the empty vector (AAV-Vec) was infused into the unilateral hippocampal DG subset of 2-month-old C57 mice for 8 weeks. A) Paradigms used for testing the cognitive functions: open field (OF) and elevated plus maze (EPM) for anxiety and depression; novel object recognition (NOR) for memory ability, Morris water maze (MWM) for spatial learning and memory, and fear conditioning (FC) for fear memory. B, C) In OF and EPM tests, all tau-overexpressing groups showed decreased duration in center or in open arm, and the decrease was more significant in Tau-WT and Tau-4Q than Tau-4R group. Data were presented as mean±SEM (N = 10∼17 each group, one-way ANOVA). *p < 0.05, **p < 0.01, versus Vec. D, E) In NOR test, both Tau-WT and Tau-4Q groups showed decreased preference toward novel object, and the decrease was more significant in Tau-4Q than Tau-WT group. No difference in distance moved among groups during NOR test. Data were presented as mean±SEM (N = 10∼17 each group, one-way ANOVA). *p < 0.05, **p < 0.01, versus Vec; ^#p < 0.05, versus Tau-4R. F) In 6 days learning test by MWM, both Tau-WT and Tau-4Q groups showed increased latency to find the hidden platform, and the increase was more significant in Tau-4Q than Tau-WT group (N = 10∼17 each group, repeated measures two-way ANOVA). *p < 0.05, Tau-4Q versus Vec, ^##p < 0.01, Tau-WT versus Vec. G–I) In MWM probe test measured at day 8 by removed the escape platform, all tau-overexpressing groups showed memory deficits compared with the empty vector controls. Data were presented as mean±SEM (N = 10∼17 each group, one-way ANOVA). *p < 0.05, **p < 0.01, versus Vec; ^##p < 0.01, versus Tau-4R. J) In FC test, both Tau-WT and Tau-4Q mice showed similarly decreased freezing time compared with the empty vector controls. Data were presented as mean±SEM (N = 10∼17 each group, one-way ANOVA). *p < 0.05, versus Vec.

DISCUSSION

Tau accumulation forming neurofibrillary tangles is hallmark pathology in the brains of AD and related tauopathies. Postmortem pathological and clinical behavioral studies demonstrate that the number of neurofibrillary tangles is positively correlated with the degree of dementia in AD patients [26, 27]. Therefore, tau accumulation plays a pivotal role in memory deterioration of AD patients. It is also observed that spatial memory loss is an early symptom in AD patients, suggesting hippocampal involvement in the early stage of AD. Postmortem studies also show that EC subset is the most susceptible region for AD patients to develop tau pathology and neurodegeneration [28]. Tau has a prion-like transmission mode and aggregation makes tau more prone to propagate. During AD onset and progression, tau starts to aggregate in EC subset in an early stage, and then the abnormal tau proteins spread to the limbic hippocampus, hippocampal subsets, and finally propagate to the whole brain [17 , 29]. Currently, the molecular mechanisms promoting tau propagation are not fully understood, and the role of tau through this route propagation in cognitive function has not been determined.

Several posttranslational modifications can promote tau aggregation, and acetylation is one of them [6 , 30]. Furthermore, a significantly increased acetylation of tau at K174, K274, K280, and K281 has been detected in the brains of AD patients and transgenic mice [4 –6] that show tau staining in the DG molecular layer at 12 months [24]. Therefore, we studied whether the acetylation of tau at these four sites increases its propagation. We first constructed hyperacetylated tau by mutating the four sites of lysine (K174, K274, K280, and K281) into glutamine (Tau-4Q). Simultaneously, the Tau-WT and Tau-4R (Lys mutated to Arg at K174, K274, K280, and K281) were respectively used as normal acetylation levels and low (or none) acetylation controls. We found that overexpressing tau including Tau-WT, Tau-4Q, and Tau-4R in EC of 2-month-old C57 mice for 3 or 6 months did not induce significant tau propagation to the DG subset. However, when the mice were reared to 13.5 months after infusion of the virus constructs, the propagation of tau from EC to DG was detected in Tau-WT and Tau-4Q but not in Tau-4R. Furthermore, Tau-WT was transmitted to the granulosa cell layer of DG while Tau-4Q spread to the inner hilus of DG, indicating that Tau-4Q (hyperacetylated) transmits farther than the Tau-WT (normal level of acetylation). These data strongly suggest that acetylation of tau at the four AD-related sites promotes a time-dependent tau propagation, which exactly fits the chronic feature of tau-related neurodegeneration as seen in AD.

Previous studies demonstrated that hippocampal overexpressing tau induced learning and memory deficits in mice [31, 32], while overexpressing tau in the EC subset of 2-month-old C57 mice for 6 months did not induce cognitive impairments [33]. Consistently, the current study also shows that overexpressing different types of tau proteins in EC for 3 and 6 months did not induce significant cognitive deficits measured by multiple behavioral paradigms, including OF, EPM, NOR, MWM, and FC tests. However, the cognitive deficit was shown when tau proteins were propagated to the hippocampal DG at 13.5 months after the infusion of Tau-WT and Tau-4Q in 2-month-old C57 mice. These data imply that only hippocampus, not EC, tau accumulation more likely impairs cognitive function. To confirm this point and further verify the effects of tau acetylation on its propagation, we overexpressed different types of tau proteins in unilateral hippocampal DG and then measured its propagation to the contralateral hippocampus and the cognitive functions of the mice. We found that Tau-WT and Tau-4Q but not Tau-4R were propagated to the contralateral hippocampus, and the propagation was more prominent in Tau-4Q than Tau-WT. These data further confirm that tau acetylation promotes its transmission. Spreading of tau to the contralateral hippocampus was also reported by upregulating P300, an acetylase [34]. Interestingly, we observed that in addition to the propagated Tau-WT and Tau-4Q, overexpressing Tau-4R in hippocampus that did not propagate also induced cognitive deficits when compared with the age-matched empty vector control. These data imply that hippocampal tau accumulation per se seems more critical than tau acetylation in impairing the cognitive functions, or human tau accumulation induces cognitive impairments through multiple mechanisms in addition to acetylation and propagation. The detailed mechanisms underlying tau-induced memory deficits, especially why Tau-4Q did not exacerbate and Tau-4R did not rescue the cognitive function deserve further investigation.

Previous studies indicate that microglial activation plays a role in tau propagation [17, 35]. In the current study, we also observed activation of microglia for 3 and 6 months and astrocytes for 6 months in the local EC region after overexpressing different types of tau proteins (Tau-WT, Tau-4Q, and Tau-4R), and there was no difference between Tau-4Q and Tau-4R. In agreement with the negative tau propagation data, glial activation in the DG subset was only detected at 13.5 months but not at 3 and 6 months after tau overexpression. These data suggest that instead of acting as an upstream stimulator, glial activation could also be a consequence of tau propagation and accumulation in the affected sub-regions. Previous studies suggest that tau can be secreted to the extracellular in the form of exosomes [18 , 37]. After secretion, it can be absorbed by surrounding cells, including neurons and glial cells [38, 39]. It will be interesting in future studies to clarify how tau propagation affects glial activity, directly engulfed by the glial cells or indirectly affected by the neighboring tau-accumulated neurons.

We also tried to measure the phosphorylation level of tau after transient transfection of different tau plasmids including Tau-WT, Tau-4Q, and Tau-4R in HEK293T cells, but no significant change was detected at Ser202 and Thr205 in different groups. According to our previous experience, we could always see increased tau phosphorylation or aggregation after overexpressing human tau in vitro or in vivo [31 , 40–44]. It will be interesting in future studies to explore whether other phosphorylation sites are involved or using different experimental models. Aβ could promote misfolding of tau demonstrated by MC1 staining [45], while the recent studies including the data shown in the current paper indicate that tau acetylation could increase its secretion or transmission [34]. Therefore, we speculate that introduction of Tau-4Q into APP/PS1 or 3xTg AD mouse models may aggravate AD-like pathologies, which may deserve further investigation. We also noticed that the brain injection sites, such as EC versus hippocampal DG, might be involved in a significantly different spatial memory presentation, but the mechanism is currently not understood.

In summary, we found in the present study tau acetylation induces a time-dependent propagation of tau from EC to DG, simultaneously glial activation in DG subset and cognitive deficits were detected in the mice.

Footnotes

ACKNOWLEDGMENTS

This work was supported in parts by the National Key R&D Program of China (2016YFC1305800), the Natural Science Foundation of China (31730035, 81721005, 91949205), and the Guangdong Provincial Key S&T Program (2018B030336001).

Authors’ disclosures available online ().

The supplementary material is available in the electronic version of this article: .

References

Grundke-Iqbal

, Iqbal

, Tung

, Quinlan

, Wisniewski

, Binder

(1986) Abnormal phosphorylation of the microtubule-associated protein tau (tau) in Alzheimer cytoskeletal pathology. Proc Natl Acad Sci U S A 83, 4913–4917.

Grinberg

, Wang

, Sohn

, Theofilas

, Sidhu

, Arevalo

, Heinsen

, Huang

, Rosen

, Miller

, Gan

, Seeley

(2013) Argyrophilic grain disease differs from other tauopathies by lacking tau acetylation. Acta Neuropathol 125, 581–593.

Goedert

, Eisenberg

, Crowther

(2017) Propagation of tau aggregates and neurodegeneration. Annu Rev Neurosci 40, 189–210.

Min

, Chen

, Tracy

, Li

, Zhou

, Wang

, Shirakawa

, Minami

, Defensor

, Mok

, Sohn

, Schilling

, Cong

, Ellerby

, Gibson

, Johnson

, Krogan

, Shamloo

, Gestwicki

, Masliah

, Verdin

, Gan

(2015) Critical role of acetylation in tau-mediated neurodegeneration and cognitive deficits. Nat Med 21, 1154–1162.

Tracy

, Sohn

, Minami

, Wang

, Min

, Li

, Zhou

, Le

, Lo

, Ponnusamy

, Cong

, Schilling

, Ellerby

, Huganir

, Gan

(2016) Acetylated tau obstructs KIBRA-mediated signaling in synaptic plasticity and promotes tauopathy-related memory loss. Neuron 90, 245–260.

Cohen

, Guo

, Hurtado

, Kwong

, Mills

, Trojanowski

, Lee

(2011) The acetylation of tau inhibits its function and promotes pathological tau aggregation. Nat Commun 2, 252.

Gorsky

, Burnouf

, Sofola-Adesakin

, Dols

, Augustin

, Weigelt

, Gronke

, Partridge

(2017) Pseudo-acetylation of multiple sites on human Tau proteins alters Tau phosphorylation and microtubule binding, and ameliorates amyloid beta toxicity. Sci Rep 7, 9984.

Min

, Cho

, Zhou

, Schroeder

, Haroutunian

, Seeley

, Huang

, Shen

, Masliah

, Mukherjee

, Meyers

, Cole

, Ott

, Gan

(2010) Acetylation of tau inhibits its degradation and contributes to tauopathy. Neuron 67, 953–966.

Fontaine

, Sabbagh

, Baker

, Martinez-Licha

, Darling

, Dickey

(2015) Cellular factors modulating the mechanism of tau protein aggregation. Cell Mol Life Sci 72, 1863–1879.

10.

Gibbons

, Lee

VMY

, Trojanowski

(2019) Mechanisms of cell-to-cell transmission of pathological tau: A review. JAMA Neurol 76, 101–108.

11.

Guo

, Lee

(2014) Cell-to-cell transmission of pathogenic proteins in neurodegenerative diseases. Nat Med 20, 130–138.

12.

Braak

, Braak

(1991) Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol 82, 239–259.

13.

Clavaguera

, Akatsu

, Fraser

, Crowther

, Frank

, Hench

, Probst

, Winkler

, Reichwald

, Staufenbiel

, Ghetti

, Goedert

, Tolnay

(2013) Brain homogenates from human tauopathies induce tau inclusions in mouse brain. Proc Natl Acad Sci U S A 110, 9535–9540.

14.

Venegas

, Kumar

, Franklin

, Dierkes

, Brinkschulte

, Tejera

, Vieira-Saecker

, Schwartz

, Santarelli

, Kummer

, Griep

, Gelpi

, Beilharz

, Riedel

, Golenbock

, Geyer

, Walter

, Latz

, Heneka

(2017) Microglia-derived ASC specks cross-seed amyloid-beta in Alzheimer’s disease. Nature 552, 355–361.

15.

Martini-Stoica

, Cole

, Swartzlander

, Chen

, Wan

, Bajaj

, Bader

, Lee

VMY

, Trojanowski

, Liu

, Sardiello

, Zheng

(2018) TFEB enhances astroglial uptake of extracellular tau species and reduces tau spreading. J Exp Med 215, 2355–2377.

16.

Hopp

, Lin

, Oakley

, Roe

, DeVos

, Hanlon

, Hyman

(2018) The role of microglia in processing and spreading of bioactive tau seeds in Alzheimer’s disease. J Neuroinflammation 15, 269.

17.

Asai

, Ikezu

, Tsunoda

, Medalla

, Luebke

, Haydar

, Wolozin

, Butovsky

, Kugler

, Ikezu

(2015) Depletion of microglia and inhibition of exosome synthesis halt tau propagation. Nat Neurosci 18, 1584–1593.

18.

Wang

, Balaji

, Kaniyappan

, Kruger

, Irsen

, Tepper

, Chandupatla

, Maetzler

, Schneider

, Mandelkow

(2017) The release and trans-synaptic transmission of Tau via exosomes. Mol Neurodegener 12, 5.

19.

Nacharaju

, Lewis

, Easson

, Yen

, Hackett

, Hutton

, Yen

(1999) Accelerated filament formation from tau protein with specific FTDP-17 missense mutations. FEBS Lett 447, 195–199.

20.

Albert

, Mairet-Coello

, Danis

, Lieger

, Caillierez

, Carrier

, Skrobala

, Landrieu

, Michel

, Schmitt

, Citron

, Downey

, Courade

, Buee

, Colin

(2019) Prevention of tau seeding and propagation by immunotherapy with a central tau epitope antibody. Brain 142, 1736–1750.

21.

Guo

, Narasimhan

, Changolkar

, He

, Stieber

, Zhang

, Gathagan

, Iba

, McBride

, Trojanowski

, Lee

(2016) Unique pathological tau conformers from Alzheimer’s brains transmit tau pathology in nontransgenic mice. J Exp Med 213, 2635–2654.

22.

Cohen

, Stackman

Jr (2015) Assessing rodent hippocampal involvement in the novel object recognition task. A review. Behav Brain Res 285, 105–117.

23.

Zhang

, Song

, Liu

, Kang

, Kwon

, Duong

, Seyfried

, Hu

, Liu

, Wang

, Cheng

, Sun

, Yu

, Levey

, Ye

(2014) Cleavage of tau by asparagine endopeptidase mediates the neurofibrillary pathology in Alzheimer’s disease. Nat Med 20, 1254–1262.

24.

de Calignon

, Polydoro

, Suarez-Calvet

, William

, Adamowicz

, Kopeikina

, Pitstick

, Sahara

, Ashe

, Carlson

, Spires-Jones

, Hyman

(2012) Propagation of tau pathology in a model of early Alzheimer’s disease. Neuron 73, 685–697.

25.

Braak

, Braak

(1995) Staging of Alzheimer’s disease-related neurofibrillary changes. Neurobiol Aging 16, 271–278; discussion 278-284.

26.

Qiu

, Jacobs

, Messer

, Salmon

, Feldman

(2019) Cognitive heterogeneity in probable Alzheimer disease: Clinical and neuropathologic features. Neurology 93, e778–e790.

27.

Thal

, Holzer

, Rub

, Waldmann

, Gunzel

, Zedlick

, Schober

(2000) Alzheimer-related tau-pathology in the perforant path target zone and in the hippocampal stratum oriens and radiatum correlates with onset and degree of dementia. Exp Neurol 163, 98–110.

28.

Jackson

(2014) Selective vulnerability to neurodegenerative disease: The curious case of prion protein. Dis Model Mech 7, 21–29.

29.

Ayers

, Giasson

, Borchelt

(2018) Prion-like spreading in tauopathies. Biol Psychiatry 83, 337–346.

30.

Cook

, Carlomagno

, Gendron

, Dunmore

, Scheffel

, Stetler

, Davis

, Dickson

, Jarpe

, DeTure

, Petrucelli

(2014) Acetylation of the KXGS motifs in tau is a critical determinant in modulation of tau aggregation and clearance. Hum Mol Genet 23, 104–116.

31.

Yin

, Gao

, Wang

, Ye

, Wu

, Fang

, Pi

, Yang

, Wang

, Lu

, Ye

, Wang

(2016) Tau accumulation induces synaptic impairment and memory deficit by calcineurin-mediated inactivation of nuclear CaMKIV/CREB signaling. Proc Natl Acad Sci U S A 113, E3773–3781.

32.

, Hong

, Wang

, Zhang

, Li

, Liu

, Sun

, Feng

, Ye

, Gao

, Ke

, Wang

, Li

, Ye

, Liu

, Wang

(2019) Tau accumulation triggers STAT1-dependent memory deficits by suppressing NMDA receptor expression. EMBO Rep 20, e47202.

33.

Chen

, Liu

, Li

, Yao

, Hao

, Dong

, Zheng

, Liu

(2018) Expression of human Tau40 in the medial entorhinal cortex impairs synaptic plasticity and associated cognitive functions in mice. Biochem Biophys Res Commun 496, 1006–1012.

34.

Chen

, Li

, Wang

, Tang

, Mok

, Tsai

, Rojas

, Karydas

, Miller

, Boxer

, Gestwicki

, Arkin

, Cuervo

, Gan

(2020) Promoting tau secretion and propagation by hyperactive p300/CBP via autophagy-lysosomal pathway in tauopathy. Mol Neurodegener 15, 2.

35.

Perea

, Llorens-Martin

, Avila

, Bolos

(2018) The role of microglia in the spread of tau: Relevance for tauopathies. Front Cell Neurosci 12, 172.

36.

Chai

, Dage

, Citron

(2012) Constitutive secretion of tau protein by an unconventional mechanism. Neurobiol Dis 48, 356–366.

37.

Simon

, Garcia-Garcia

, Royo

, Falcon-Perez

, Avila

(2012) Proteostasis of tau. Tau overexpression results in its secretion via membrane vesicles. FEBS Lett 586, 47–54.

38.

Clavaguera

, Bolmont

, Crowther

, Abramowski

, Frank

, Probst

, Fraser

, Stalder

, Beibel

, Staufenbiel

, Jucker

, Goedert

, Tolnay

(2009) Transmission and spreading of tauopathy in transgenic mouse brain. Nat Cell Biol 11, 909–913.

39.

Frost

, Jacks

, Diamond

(2009) Propagation of tau misfolding from the outside to the inside of a cell. J Biol Chem 284, 12845–12852.

40.

Wei

, Ye

, Wang

, Zhu

, Hu

, Wang

, Ke

, Tian

, Wang

(2018) Tau-induced Ca(2+)/calmodulin-dependent protein kinase-IV activation aggravates nuclear tau hyperphosphorylation. Neurosci Bull 34, 261–269.

41.

, Yin

, Liu

, Fang

, Tao

, Wei

, Zuo

, Yin

, Ke

, Wang

(2020) Tau inhibits PKA by nuclear proteasome-dependent PKAR2alpha elevation with suppressed CREB/GluA1 phosphorylation. Aging Cell 19, e13055.

42.

Liu

, Zhou

, Xie

, Li

, Ye

, Li

, Ke

, Wang

, Xu

, Li

, Yang

, Liu

, Wang

, Li

, Wang

(2020) Tau acetylates and stabilizes beta-catenin thereby promoting cell survival. EMBO Rep 21, e48328.

43.

Feng

, Luo

, Zhang

, Yang

, Hong

, Sun

, Li

, Hu

, Li

, Zhang

, Li

, Yang

, Wang

, Liu

, Wang

(2020) MAPT/Tau accumulation represses autophagy flux by disrupting IST1-regulated ESCRT-III complex formation: A vicious cycle in Alzheimer neurodegeneration. Autophagy 16, 641–658.

44.

Zheng

, Li

, Tian

, Liu

, Wang

, Yin

, Yue

, Ma

, Wan

, Wang

(2020) Interneuron accumulation of phosphorylated tau impairs adult hippocampal neurogenesis by suppressing GABAergic transmission. Cell Stem Cell 26, 331–345e336.

45.

Heraud

, Goufak

, Ando

, Leroy

, Suain

, Yilmaz

, De Decker

, Authelet

, Laporte

, Octave

, Brion

(2014) Increased misfolding and truncation of tau in APP/PS1/tau transgenic mice compared to mutant tau mice. Neurobiol Dis 62, 100–112.